Masuda Isaeva, Asakabank, Uzbekistan

Uzbekistan is among the top remittance-receiving countries in the world. However, there is a lack of empirical research on how remittances influence recipients' financial inclusion and spending patterns. Existing studies primarily rely on survey-based or macroeconomic data, which often overlook key characteristics of remittances and their recipients. As a result, these studies reveal a homogeneous effect of remittances and fail to provide practical insights for financial institutions. This study addresses this research gap by exploring the factors shaping saving behaviour among remittance recipients in Uzbekistan. Using a dataset from 2017 to 2023, machine learning models, and SHAP values, we investigate the heterogeneous contribution of characteristics of recipients and remittances on the adoption of bank deposits. Our results reveal that gender, age, amount, and duration of remittances significantly influence deposit acceptance, while the origin of remittances has no effect. The insights derived from this study can aid policymakers and financial institutions in designing targeted strategies to encourage saving among remittance recipients and enhancing financial inclusion in country.

Bank deposits, machine-learning (ML) models, remittances, SHAPLEY values.